In 1984, Joehl et al. (1) reported that IM injected meperidine and morphine delayed the passage of dye into the duodenum in healthy volunteers for slightly less than 2 h and 4 h, respectively, and that butorphanol and nalbuphine did so for approximately 3 h. Because of its short duration of effect, remifentanil is usually given via an IV infusion. It is also used as a component of monitored anesthesia care. Therefore, it may cause abdominal pain if given to a patient who has cholecystitis and/or cholelithiasis by causing spasm of the sphincter of Oddi. Other opioids have caused acute abdominal pain in such patients. Dosing guidelines (2) suggest an infusion rate of 0.025–0.2 μg · kg−1 · min−1 in a patient who is awake and breathing spontaneously and 0.2–0.5 μg · kg−1 · min−1 as part of a general anesthetic technique. The effect of remifentanil on the liver, gall bladder, bile ducts, or sphincter of Oddi is unknown, although it should act like other μ agonist opioids. Previous investigators have shown that morphine, meperidine, fentanyl, butorphanol, and nalbuphine increase sphincter of Oddi tone, delaying the passage of dye into the duodenum (1) or increasing pressure in the common bile duct (1,3–5) or both. However, one of these studies resulted in sphincter of Oddi spasm in only 3 of 100 patients receiving up to 10 μg/kg of fentanyl as a component of general anesthesia (5). Possibly, the effect of remifentanil on the sphincter of Oddi, if it indeed causes spasm, could dissipate rapidly when remifentanil is no longer given.
In this study, we examined the effect of a remifentanil infusion (0.1 μg · kg−1 · min−1) on the time for passage of a radioactive technetium-labeled dye from IV injection until its presence in the duodenum in healthy volunteers (hepatobiliary scan). The study goals were 1) to see if remifentanil delayed the time of a control hepatobiliary scan, 2) to measure that delay, if it occurred, and 3) to determine what the recovery time would be for dye to appear in the duodenum after the remifentanil infusion was stopped.
Six healthy volunteers participated in this institutionally approved study after they signed an informed consent. A power analysis showed that six subjects were required to show an 8-min or greater delay in transit time of dye from control values with an 88% reliability at a significance of <0.05 (Sigmastat statistical software; HALLoGRAM, Aurora, CO). The volunteers had control hepatobiliary imaging performed in the nuclear medicine department after a 6-h fast. Before this, their medical history was recorded, a physical examination was performed, and liver function tests were performed. To participate, volunteers could have no prior or current gall bladder or liver disease, no history of alcohol or drug abuse, no current use of any drugs that could interfere with the scans or interact with remifentanil, and normal liver function tests. The hepatobiliary scan was performed using 5 mCi of a radioactive technetium-labeled derivative of iminodiacetic acid (99mTc-IDA) injected IV (1). The abdomen was then imaged to look for activity in the duodenum. The time from injection to initial appearance of activity in the duodenum was recorded as the control time.
After 2 wk, each volunteer returned to the nuclear medicine department after a 6-h fast and no change in their health. A blood pressure cuff and pulse oximeter, from which a heart rate was determined, were attached and control values obtained. An infusion of normal saline was started. An infusion of remifentanil was delivered from an infusion pump into the tubing of the IV infusion. Remifentanil was diluted to a concentration of 50 μg/mL and infused at a rate of 0.1 μg · kg−1 · min−1 for 30 min (approximately three half-lives) to achieve close to a steady state level of remifentanil in the blood and at the receptor sites. An injection of the same dose of the dye that was given at the control session was then administered IV. The remifentanil infusion was continued at the same rate for the length of the control scan plus 10 min. This time was chosen a priori, as it seemed that if remifentanil caused spasm of the sphincter of Oddi, that it would do so as long as the infusion was running. If dye appeared in the duodenum before this time, then remifentanil did not delay the passage of dye into the duodenum and, thus, caused no sphincter of Oddi spasm. The infusion of remifentanil was then stopped, and the time that the dye appeared in the duodenum was noted. The duration of time from dye injection until it appeared in the duodenum between the control and remifentanil treatment sessions was compared by using paired Student’s t-tests, in which P < 0.05 was considered statistically significant. The time from the end of the remifentanil infusion until dye appeared in the duodenum was considered the recovery time and is descriptive. Figure 1 shows a representative time line of the events of this study. The time from D to C would be the control time determined in the first study session. It is compared with the time from D to A that was determined in the second study session.
Volunteers were monitored from before the remifentanil infusion until they recovered from remifentanil’s effects. If there were untoward cardiovascular effects, it was planned to decrease or stop the remifentanil infusion. If respiratory depression occurred, volunteers were first encouraged to take deep breaths; if that was not sufficient, then nasal oxygen was applied. If further treatment was necessary, the infusion rate of remifentanil could be decreased or stopped. When the volunteers had no signs of respiratory or cardiovascular depression and could sit and stand without dizziness or other opioid effects, they were sent home with another person.
A member of the nuclear medicine department who did not know which treatment session they represented read the hepatobiliary scans.
The six volunteers were four women and two men, 26 to 49 yr old. Five of the volunteers weighed between 63 and 80 kg, and one man weighed 104 kg. All were within 35% of ideal body weight for their height.
No volunteer required any remifentanil rate adjustment for cardiovascular or respiratory depression. Three of the volunteers needed encouragement to breathe deeply, and one required nasal oxygen to maintain oxygen saturation at 95% or above. All volunteers became drowsy, and some slept for awhile during the remifentanil infusion. All volunteers were capable of leaving the hospital within 30 min after the scanning time, which was between 35 and 60 min from stopping the remifentanil infusion. None of the volunteers experienced nausea or vomiting, but they all had mild pruritus. Thus, all volunteers experienced definite opioid effects from remifentanil.
The control scan time was 20.5 ± 9.9 min (range 10–33 min). This is within the normal range for this test. The total time from dye injection until it appeared in the duodenum under the influence of remifentanil was 50.3 ± 17.3 min (range 30–81 min). This was significantly longer (P < 0.002) than the control values (Table 1). No dye appeared in the duodenum during the remifentanil infusion in any volunteer.
The recovery time for dye to appear in the duodenum after stopping the remifentanil infusion was 19.8 ± 12.4 min (range 5–40 min). All times are mean ± SD. The recovery times in our six volunteers were 5, 9, 19, 22, 24, and 40 min. Figure 2 shows a typical example of control and treatment scans from one volunteer.
This study showed that remifentanil significantly prolonged the time from dye injection until it could be detected in the duodenum, presumably by increasing sphincter of Oddi tone. There was no dye in the duodenum during the remifentanil infusion, showing that an infusion rate of 0.1 μg · kg−1 · min−1 can prevent dye from entering the duodenum. After the infusion stopped and the drug level was decreasing, residual remifentanil effects were variable in time. With the spread of recovery times seen in this study, individual circumstances that are present during gall bladder surgery would dictate whether remifentanil would have any advantage as the opioid of choice as a component of the anesthetic technique. If the opioid effect on the sphincter of Oddi were gone shortly after the infusion stopped, the drug effect would not complicate the diagnosis if obstruction at the sphincter of Oddi were seen on intraoperative cholangiography, differentiating drug effect from a nonopaque stone. However, this study was not performed on patients undergoing biliary tract surgery, and common duct pressures were not measured. Therefore, we can only conclude that remifentanil delayed the time from injection of dye until it was observed in the duodenum. Furthermore, the duration of this effect when remifentanil administration is stopped is less than that which occurred after other opioids, such as morphine and meperidine, when they were injected IM (1).
The effects of remifentanil appear to decrease rapidly after a remifentanil infusion is stopped. The exact time could depend on the dose and duration of the infusion and what other drugs are given concurrently. Glaxo-Wellcome, the manufacturer of remifentanil, suggested that an infusion rate of 0.1 μg · kg−1 · min−1 is the highest infusion rate that would not cause unacceptable respiratory depression in awake, spontaneously breathing patients (2). This might be an appropriate dose for monitored anesthesia care or postoperative analgesia. However, even this dose can cause respiratory depression in some people, as was seen in this study.
Remifentanil is much more potent than meperidine and morphine and equipotent to fentanyl. However, trying to compare the dose of remifentanil we used with the doses of other narcotics used in the previous studies listed in the bibliography is not possible. Remifentanil was given by infusion (as it is used clinically), and the other drugs were given either as an IM injection or single or repeated IV doses during general anesthesia. The plasma concentration of the various opioid analgesics at the time either dye was injected or biliary pressures measured is unknown. Therefore, we cannot say that the remifentanil effects on the sphincter of Oddi in this study are comparable to the effects of equipotent doses of other drugs. However, the remifentanil dose we used was the largest safe dose we assumed we could administer to volunteers.
Injecting naloxone (6) or glucagon (7) can counteract the effect of opioids at the sphincter of Oddi. However, it would be better if the opioid effect on the sphincter of Oddi were allowed to wear off spontaneously rather than injecting naloxone or glucagon. Naloxone would also reverse the positive analgesic effect, and glucagon, although it rarely causes hypersensitivity reactions, causes hyperglycemia and can cause nausea in conscious patients. This study showed that the passage of dye into the duodenum occurred, on average, approximately 20 minutes after a remifentanil infusion at a rate of 0.1 μg · kg−1 · min−1 was stopped.
1. Joehl RJ, Koch KL, Nahrwold DL. Opioid drugs cause bile duct obstruction during hepatobiliary scans. Am J Surg 1984; 147: 134–8.
2. Remifentanil [package insert]. Research Triangle Park: Glaxo Wellcome, Inc., 1998.
3. Radney PA, Duncalf D, Novakovic M, Lesser ML. Common bile duct pressure changes after fentanyl, morphine, meperidine, butorphanol, and naloxone. Anesth Analg 1984; 63: 441–4.
4. McCammon RL, Stoelting RK, Makura JA. Effects of butorphanol, nalbuphine, and fentanyl on intrabiliary tract dynamics. Anesth Analg 1984; 63: 139–42.
5. Jones RM, Detmer M, Hill AB, et al. Incidence of choledochoduodenal sphincter spasm during fentanyl-supplemented anesthesia. Anesth Analg 1981; 60: 638–40.
6. McCammon RL, Viegas OJ, Stoelting RK, Dryden GE. Naloxone reversal of choledochal sphincter spasm associated with narcotic administration. Anesthesiology 1978; 48: 437.
7. Jones RM, Fiddian-Green R, Knight PR. Narcotic-induced choledochoduodenal sphincter spasm reversed by glucagon. Anesth Analg 1980; 59: 946–7.